Now we're investigating the transmembrane variant, obviously, but this has been the biomarker for prostate cancer serum biomarker before PSA was available. And this is a graph showing the spike of publications somewhere in the '80s and '90s when actually PSA was discovered and was approved. And then you see how the publication, its interest on ACP3 really goes down. Then we had the approval of Sipuleucel-T, which led to another spike. And after that, it felt like the community didn't have huge interest in ACP3 as a target. And that really changed when the company Philochem discovered this new ligand OncoACP3-DOTA targeting ACP3 with very high affinity and very high selectivity. So this was enabled by Samuel Cazzamalli and Dario Neri and their teams. And this was eventually bought by RayzeBio. So we got the ligand in our hands in 2024 and we established radiosynthesis, performed small-animal imaging and were very convinced by these images.
So we decided to translate this to patients. You've seen this patient already in the slides from Jeremy. That's a very nice patient that illustrates nicely the biodistribution that we see for Gallium-68-OncoACP3-DOTA here compared to PSMA-1007. So essentially we don't see any uptake in the salivary glands, lacrimal glands, a lot less, poor uptake in the liver, spleen and kidneys. And we see somewhat of a higher retention in the blood pool, but it's kind of an unfair comparison because these first 25 patients were really one hour post-injection compared to two-hour post-injection PSMA-1007. This patient had a single metastasis that had an SUV of 105 in the PSMA PET and SUV of 210 in the ACP3 PET. So we're definitely playing in the ballpark where Dr. Hofman mentioned we have this transformative potential for radioligand therapy with this tracer.
So in our first publication, we compare 25 patients in regard which tracer would be better. Of course, it's a hard thing to say, but we set this metric as one of the traces in an individual patient either having 50% more uptake consistently in the metastases or displaying 10% more lesions. And looking at these 25 patients in a quite mixed cohorts of biochemical recurrence, metastatic prostate cancer and post-Pluvicto, we found that in 11 patients, OncoACP3 showed a better performance compared to PSMA-1007. A number of the patients that we scanned, we scanned them in order to later to evaluate them for radioligand therapy targeting ACP3. And those patients, we scan dynamically to understand the dynamics of this tracer in order to better understand the potential for radioligand therapy. And now just two days ago, accepted publication in JNM, we investigated this dynamics to establish pharmacokinetic modeling with OncoACP3.
So we found that a two-compartment exchange model with irreversible binding is really best to describe these kinetics. And we could perform dosimetry together with Julia Brosch-Lenz who's now in Augsburg, Germany who established the dose for Gallium-68-OncoACP3-DOTA, which basically plays in a similar ballpark as a PSMA agent, maybe with a mostly slightly lower dose. And what is striking here, what you may see in this patient is that the tumor uptake over time is increasing quite a bit from one hour to two hours. And this is something that we observed quite consistently. If you look at the graph on the right-hand side showing the stark increase in absolute SUV in the lymph node and osseous metastases. And we also observed that this somewhat higher blood uptake is actually going down quite nicely on the course of hours.
So in some of these patients, we had additional blood withdrawals up to six hours after injection of the gallium probe, and we could fit this data nicely and concluded that this will not be very bothersome for radioligand therapy because of course this was our first concern, seeing the somewhat higher blood retention that this would translate to higher bone marrow doses, but looking at this kinetics, it appeared quite unlikely.
So looking at these very favorable kinetics in terms of also doing later scanning time points and getting a Siemens Vision Quadra scanner in November last year, we actually switched to two hour time point imaging. And this data set is not fully analyzed yet, but you could imagine based on the dynamics that we show, we have kind of shifted the head-to-head comparison with PSMA further towards the direction of OncoACP3 for diagnostic PET imaging.
Coming to radioligand therapy with lutetium, I think this is a very nice slide that illustrates quite well the potential that this ligand may have in prostate cancer. On the left side, you see three patients that we treated in compassionate use at our center as first-in-man with lutetium-OncoACP3-DOTA here on the very left, you see the corresponding PET images at baseline and you see that we tuned the PET images as well as the longitudinal SPECT images, two hour, one day, two day or seven to eight days post-injection to the same SUVs. And on the right-hand side, you see the same for the theranostic pair of PSMA-1007 and PSMA-617. And so what you see here, there's no magic involved for both of these theranostic pairs. What you see in PET is essentially what you get in SPECT later on. So when you compare the baseline PET to the two-day SPECT, you definitely see a quantitative correspondence.
It appears to be somewhat similar for the PSMA theranostic pair as well as OncoACP3. And it gets very interesting if you compare the seven to eight time point to the two-day time point for these two different ligands. And you see that for OncoACP3-DOTA, this is essentially very stable. It's hard to even see, to even spot a difference of metastatic uptake between two days and seven to eight days. And essentially the effective retention time that we calculated matched the physical decay time of lutetium in a number of these metastases. And when you compare that to PSMA-617, it's quite different. So you easily see by eye how the uptake from two days to seven to eight dose goes down quite a little bit compared to OncoACP3. So taking this very fantastic, I believe, dynamic properties of OncoACP3 on the scale of days, together with absence of salivary gland uptake and very clean background and very high SUVs that we see in both PET and SPECT, I think it's a very interesting delivery strategy for both lutetium, but of course also for longer-lived alpha emitters such as actinium.
These are the results from the first four patients that we treated with OncoACP3. This is of course a difficult setting. This is really last line compassionate use and all these patients have seen Pluvicto already and most of them have progressed on it. So this is of course a tough task to challenge them again with lutetium, although having a different delivery strategy. However, we did so quite successfully in some of these patients, particularly the first patient here on the left side was actually the first patient that we treated with 2.55 and 4.1 GBq in three cycles and basically took the PSA down to a level where it took it more than 200 days to get back to the baseline. And this was a patient that have seen 11 cycles of Pluvicto upfront.
For the two patients on the right side, you see that they were progressing. They both had liver metastasis. They were in an aggressive state of the disease, so we couldn't really catch them with OncoACP3. And the second patient, we probably had some sort of delayed dose. So I felt that this patient probably still profited from the treatment. Looking at the doses here on the bottom of this table, you see when we look at the bone marrow dose, it's in a similar ballpark as we've seen it in the Vision trial. So it doesn't seem so far that we get more bone marrow dose and we don't really see that we have more bone marrow toxicity than we would expect for Pluvicto treatment. And pretty much it looks like for all other organs, the dose actually appears to be lower than for Pluvicto.
And with that, thank you very much again for the invitation to my collaboration partners for these papers, particularly to Florian, Konstantin, Julia, and my boss, Michael Schäfers, and to the team of RayzeBio and Philochem for enabling this. Thank you very-